Networker

Re: [Networker] /nsr/index Cleanup?

2010-03-24 10:32:30
Subject: Re: [Networker] /nsr/index Cleanup?
From: George Sinclair <George.Sinclair AT NOAA DOT GOV>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Wed, 24 Mar 2010 10:29:43 -0400
tbirkenbach wrote:
I'm unclear on why you would want to remove the client's media database entries 
using nsrmm? Why not just leave those in there and just get rid of the 
/nsr/index directory itself and the resource, too?


Most if not all the system that I remove from backup are dead and gone.  In my (maybe 
somewhat vague) attempt at trying to keep a clean backup system, I try and purge all 
non-active and non-relevant data.  You're right in that IF a client needed to perform a 
recovery for a "dead and gone" system, then this process would introduce 
unnecessary challenges had those references been left in place.  I forewarn my customers 
telling that once they tell me to remove the system from backup, after four weeks time 
(our standard retention time) all references are purged, unless they request a longer 
retention period.

Maybe it's overkill.  Maybe it doesn't work the way I think it does, but this 
is the info and process I collected from this forum when I asked a similar 
question not too long ago.  If there's a better way, I'm happy to change my 
procedures.  That's why I threw this info back out here, 1) to share what I've 
collected, and 2) to see what others think.  I appreciate the feedback George!

Well, if they're willing to accept that risk then you have nothing to be concerned about. One thing you could entertain, however - and this is really bending over backwards to be a nice backup admin - would be to create a listing (ASCII text; mminfo) of the media database entries for every client you purge and then store that somewhere for each affected client. I can't imagine that would take up very much space. It could also be backed up itself. Then, if in the future, someone needed to recover something, you could always use scanner, and you could as least find which tape, mediafile and mediarec contained the required save set. That would at least allow you to get to the part of the tape where the save set started, obviating the need to read through the whole volume. Regardless, however, without those listings, you wouldn't know which tape the save set(s) was on.

Of course, you could always just set up a 30 day eval copy on another server and recover an older copy of the media db and then recover the data from there, but this is all beyond the purview of your requirements as you outlined them. Maybe it's just as well, too, that they've accepted that point of no return as that makes your job easier and keeps everything simple. There is elegance in simplicity!

George


+----------------------------------------------------------------------
|This was sent by tom.birkenbach AT wmich DOT edu via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER



--
George Sinclair
Voice: (301) 713-3284 x210
- The preceding message is personal and does not reflect any official or unofficial position of the United States Department of Commerce -
- Any opinions expressed in this message are NOT those of the US Govt. -

To sign off this list, send email to listserv AT listserv.temple DOT edu and type 
"signoff networker" in the body of the email. Please write to networker-request 
AT listserv.temple DOT edu if you have any problems with this list. You can access the 
archives at http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER

<Prev in Thread] Current Thread [Next in Thread>